A NOVEL AI ASSISTED DISEASES DETECTION ON FRUITS AND GRADING OF FRUITS FOR HEURISTIC BASED SMART FARMING

Abstract

For any Developing country like India agriculture plays an important role and contribute major part of income to the country, so there is a need to grow and increase the yield effectively. In order to achieve above objectives, one has to monitor on the diseases stating from plantation to harvesting. This investigation made an attempt to use Artificial Immune System (AIS) and image processing to (i) Identify the diseases on fruits like Grapes and Apple (ii) Grading of fruits. Disease identification aims different features like color, texture and shapes. Which are considered as feature vectors in this work. To extract color feature HSV histogram value concept is used, for texture wavelet transform method and for shape morphology methods are used. After extracting the above said features we used AIS algorithm as a classifier to classify the diseases and it is observed that the color and morphology shows better results than texture. Grading of fruit aims fruit segmentation which calculates healthy and infected portion of fruit. At the end we practically implemented AIS algorithm and results are obtained from MATLAB.

Specification

Description:In this smart farming frame work (100) we developed a image processing model (101) and proposed a Artificial immune algorithm (102) considering color, morphology and texture as a feature vectors to extract images for learning data base. and investigated and supervised diseases on fruits (103) and grading of fruits (104) like grapes and an apple and propose a solution to get better productivity.
The feature vectors namely color-morphology-texture are worn to extract images for learning data base.One of the important features that human uses for object discrimination. The principle of color image processing consists of namely (i) Color mapping (ii) Individual color planes processing (iii) Color vector processing. Here, HSI color space is used to capture the image. , Claims:1. An AI assisted diseases detection on fruits (100) comprising: a design using mathematical modeling to improve the productivity;
2. An AI assisted diseases detection on fruits (100) of claim 1, wherein the designed mathematical model (101) is proposed.;
3. An AI assisted diseases detection on fruits (100) of claim 1 and claim 2, wherein the heuristic-based Artificial Immune algorithm (102) is proposed to as a classifier to classify the diseases;
4. An AI assisted diseases detection on fruits (100) of claim 1, claim 2 and claim 3, wherein Image processing technique (104) is used to grading of fruits using K-clustering;and
5.An AI assisted diseases detection on fruits (100) of claim 1, claim 2 and claim 3 and claim 4, wherein the proposed algorithm shows that the color and morphology shows better results than texture in improving the productivity (105).

Documents